W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2009

Re: Warn header (P6) vs RFC2047 encoding

From: Henrik Nordstrom <henrik@henriknordstrom.net>
Date: Thu, 23 Jul 2009 22:51:26 +0200
To: Julian Reschke <julian.reschke@gmx.de>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
Message-Id: <1248382286.14420.121.camel@localhost.localdomain>
tor 2009-07-23 klockan 22:14 +0200 skrev Julian Reschke:

> So do we have evidence that (a) RFC2047-encoded Warn headers are in use, 
> and (b) that UAs do support them?

No on both. We rather have evidence that most UAs blindly ignores
Warning not yet having implemented this part of the specs, and that the
ones who tries to display the warning do not support RFC2047.
Additinally most servers don't even issue Warning headers either even
when they should.

Brain cells firing again after reading over the Warning text a couple of
times more. And change of opinion. I am fine with reducing Warning
warn-text to the same level as Reason Phrase. English US-ASCII intended
mainly for debugging/tracing with the actual error description being the
code and not the text. Additionally this enables UAs to show the
warnings in various ways as they see fit, not restricted to textual
display.

This works fine for all the defined errors except the "misc" errors (199
& 299).

Regards
Henrik
Received on Thursday, 23 July 2009 20:52:08 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:08 GMT